The same factors that are expected to keep air masses clashing with support from a strong jet stream keep the risk of major severe weather outbreaks in the nation very real between now and May, with above-normal activity anticipated for the Lower Midwest, Mid-South and Tennessee Valley in particular. Warmer than normal Gulf of Mexico waters will provide more fuel through greater evaporation, while the La Nina chill from the North keeps the baroclinic energy index high in between the Northern and Southern states. However, the lack of snow cover in the lower 48, if it persists, could shift the focus of the biggest outbreaks further North and West than last spring. Last winter and spring had a strong La Nina this winter and spring we have a barely moderate and weakening La Nina in the Pacific.

The term "Dixie Alley" was coined in 1971. The first statistical analysis verifying the validity of "Dixie Tornado Alley" (DTA) was provided in 2003 with additional research in 2007, 2010-2011. The more familiar "Tornado Alley" is in the Plains states: Plains tornado alley (PTA).

In DTA the risk of a strong or killer tornado at the peak is substantially less than at the peak in PTA. But the DTA can be seen as a more persistent, moderate risk through much of the year rather than an extreme risk during a more confined "season" as is observed in the PTA. However, DTA has a greater risk of of killer tornadoes overall than PTA. Dixie alley has about a 50% greater risk of strong tornadoes with over a third occurring between 9pm and 7am. Another concern is cool season tornadoes tend to move much faster reducing time for people to get to shelter.

Between 1986 and 2005 75% of tornado fatalities between the two alleys occurred in DTA! Statistics also show that the probability of detection (POD) and lead time for strong tornadoes are BOTH MUCH LESS less in Dixie Alley, (POD of 80 vs. 68% and 14 minutes vs. 10 min). This means for Dixie Alley more tornadoes undetected or detected too late and less advance warning time for those that are detected. Hence southern tornadoes can be more dangerous.

A 2010 research paper by Michael Frates, a graduate assistant at the University of Akron, analyzed the occurrence of strong F3 to F5 tornadoes with tracks greater than 20 miles from 1950 to 2006.

In the study, which was presented at the annual meeting of the American Association of Geographers in Washington, Frates concluded that Dixie Alley, an area stretching from eastern Texas across parts of Missouri, Arkansas, Louisiana, Mississippi, Tennessee and Alabama into northeast Georgia and including most of metro Atlanta (mainly NW), had the greatest frequency of strong, long-track tornadoes.

The National Weather Service reports that from 1950 to 2008, Cobb and Hall counties recorded the most tornadoes in north Georgia, with 27 twisters in each county. Other metro Atlanta counties high on the list include Fulton, with 23 tornadoes during that period, Bartow with 22 and Cherokee and Carroll with 21 each. In the south part of the state Worth County in southwest Georgia recorded 30 tornadoes in that time while Colquitt County was struck by 28.

A new study last year led by meteorologist Grady Dixon of Mississippi State University published in the Bulletin of the American Meteorological Society found residents of the so-called Dixie Alley may witness the most tornadoes, since tornadoes tend to be on the ground longer in the South. In fact, the research found that the most tornado-prone area in the country is Smith County, in southeastern Mississippi.

By focusing on the path length of past tornadoes — not just the sheer number of tornadoes but how long they stay on the ground — Dixon and his study co-authors found that there is a greater likelihood of tornadoes hitting parts of Dixie Alley than Tornado Alley.

The reason tornadoes tend to be on the ground longer in the Southeast, Dixon says, is not so much based on geography or topography but the fact that the Southeast tends to see tornadoes in early spring, when the atmosphere spawns more long-lasting and faster-moving tornadoes. By the time the Plains gets its active tornado season in May, he says the atmosphere isn't as volatile, so the tornadoes often aren't as long-lived or as fast-moving.

What about a Georgia tornado alley?

Data since 1950 show that there is a kind of mini  tornado alley in a corridor west, north, and northeast of the city running roughly along and "Northwest" of I-85. These counties have had both the most and the strongest tornadoes in the metro area.

In all these studies it's worth noting that the nature of the data looked at and the methodology chosen can impact the patterns revealed, known in science as a data collection artifact. Improved radar technology also increases detection to boost numbers.

It is also important to remember that the numbers per county are based on CONFIRMED tornado touchdowns, such as those citizens call in to report seeing, storm chasers see, or damage that is found and confirmed later. So in more populated counties you're going to have higher numbers because there are so many more people to spot and report one, and more densely packed areas where damage can be recorded.  In very rural or forested counties, many more tornadoes may actually touch down, but because they're so sparsely populated, no one sees them to report it, and there's no recordable damage. So some of the mini-alley in Georgia may be a consequence of these counties having a higher population than others. But that may not be the full explanation, as some of the adjacent counties to the east of I-85 also have high populations. New close up satellite imagery should help correct this weakness in the data going forward.

More on "Dixie Alley"

The idea that there is a Dixie "leg" to traditional "Tornado Alley" has been around informally for a couple decades, and formally in the literature since at least 2007. Remember, what a tornado occurrence map looks likes depends on the criteria, data, and methodology used in the research study. The latest study was presented at the annual meeting of the American Association of Geography, of which I am a member.

Tornadoes are commonly thought to focus their destruction on "Tornado Alley," a stretch of land that spans Texas and Oklahoma North toward Minnesota and Illinois. It is a term from the media not from meteorologists. The new research finds that tornadoes actually have several other preferential zones across the eastern United States.

The Frates study found three other "alleys," one of which is even busier than the original.

Earlier studies have analyzed tornado distribution by county. Frates instead divvied up the map into grid cells all of the same size, a method that he says provides greater resolution and catches more of the nuances of geography.

Dixie Alley, centered over Mississippi and northern Alabama, beat out Tornado Alley in tornado number, Frates found. Over the 56 years examined, an average of 2.92 tornadoes hit each cell in Dixie Alley, followed by an average of 2.59 tornadoes in each cell of Tornado Alley. Second and third runners-up are Hoosier Alley (Indiana) and Carolina Alley (centered over both Carolinas).

The two top tornado-prone regions have different storm patterns: Tornadoes tend to strike Dixie Alley all year round, while Tornado Alley's peak season is only four months long. Frates would like to incorporate data such as the number of deaths and the amount of property damage caused by tornadoes in the two regions, particularly since warning systems aren't as advanced in Dixie Alley as they are in Tornado Alley. He hopes the new study will make people "more aware that more tornadoes occur in areas other than Tornado Alley." It is significant that tornadoes in "Dixie Alley" tend to be strong and long-lived on the ground for long stretches, and that they are often hidden by rain and/or occur at night. These factors all add to their deadly potential in the South.

Tornadoes spawned by landfalling hurricanes are NOT part of the Dixie leg of traditional tornado alley. Twisters from hurricanes tend to be much weaker, smaller and briefer-- on the ground for short periods of time, and are not usually EF 3 or higher on the Enhanced Fujita tornado scale. The tornado alley/Dixie alley tornadoes are more likely to be the more destructive EF-3 to EF-5 variety.